An oral pill cut lipoprotein(a) up to 85.8% — with no evidence yet on events

Lipoprotein(a) — Lp(a) — is an inherited cholesterol-carrying particle that raises the risk of heart attack, stroke and narrowed aortic valves, and diet and statins barely touch it. That has made it one of cardiology's most pursued targets. Muvalaplin is unusual among the drugs chasing it: a once-daily pill, rather than an injection, that blocks the particle from assembling in the first place. KRAKEN, published in JAMA, is the phase 2 test of whether that effect holds over three months in people at high cardiovascular risk [s1].

The result is a very large fall in Lp(a). What KRAKEN cannot tell you is whether that fall prevents a single heart attack — and that gap is the whole story of where this drug stands.

What KRAKEN did

KRAKEN was a phase 2, placebo-controlled, randomised, double-blind trial that enrolled 233 participants with lipoprotein(a) concentrations of 175 nmol/L or greater who also had atherosclerotic cardiovascular disease, diabetes, or familial hypercholesterolemia [s1]. It ran at 43 sites in Asia, Europe, Australia, Brazil, and the United States between December 10, 2022, and November 22, 2023 [s1]. Participants were randomly assigned to orally administered muvalaplin at 10 mg/d (n=34), 60 mg/d (n=64), or 240 mg/d (n=68), or to placebo (n=67), for 12 weeks [s1].

The median age was 66 years, 33% were female, and 27% identified as Asian, 4% as Black, and 66% as White [s1]. The trial is registered as NCT05563246, with Eli Lilly and Company as the sponsor, and is complete [s2].

One design detail matters more than it looks. Lp(a) is normally measured with an assay that can be thrown off by the very mechanism muvalaplin uses, so the trial reported its primary endpoint two ways: with a newer assay that measures intact Lp(a) particles, and with a traditional apolipoprotein(a)-based assay [s1]. The two give different numbers, and honest reporting shows both.

The numbers

On the intact-particle assay, muvalaplin produced placebo-adjusted reductions in lipoprotein(a) of 47.6% (95% CI, 35.1%-57.7%) at 10 mg/d, 81.7% (95% CI, 78.1%-84.6%) at 60 mg/d, and 85.8% (95% CI, 83.1%-88.0%) at 240 mg/d [s1]. On the older apolipoprotein(a)-based assay the reductions were smaller — 40.4% (95% CI, 28.3%-50.5%), 70.0% (95% CI, 65.0%-74.2%), and 68.9% (95% CI, 63.8%-73.3%) across the same three doses [s1].

The drug also nudged a second atherogenic marker: dose-dependent reductions in apolipoprotein B of 8.9% (95% CI, -2.2% to 18.8%), 13.1% (95% CI, 4.4%-20.9%), and 16.1% (95% CI, 7.8%-23.7%) [s1]. There was no change in high-sensitivity C-reactive protein, a marker of inflammation [s1]. No safety or tolerability concerns were observed at any dosage over the 12 weeks [s1]. An earlier 14-day phase 1 study had already shown muvalaplin reduced Lp(a) up to 65% and was well tolerated [s1].

What a big surrogate number does and does not buy

A placebo-adjusted reduction above 80% is among the largest reported for any Lp(a)-lowering agent, and the fact that it comes from a pill rather than a periodic injection is what makes muvalaplin distinctive. But Lp(a) concentration is a surrogate — a laboratory stand-in for the thing patients and doctors actually care about, which is whether lowering it prevents cardiovascular events.

That link is not yet established for any Lp(a) drug. The entire class rests on genetic and epidemiological evidence that lifelong high Lp(a) tracks with more heart disease, plus the assumption that lowering it pharmacologically will help. Whether it does, and by how much, is the question that large outcome trials of injectable Lp(a) agents are still working to answer. KRAKEN adds nothing on that front, because it was neither designed nor long enough to measure heart attacks, strokes or deaths — as the investigators state plainly, the effect of muvalaplin on cardiovascular events requires further investigation [s1].

Two further cautions belong next to the headline figure. The trial was funded by the company developing the drug [s2], the standard caveat for an industry-run phase 2 study. And the two assays disagree by roughly 15 percentage points at the top doses [s1] — a reminder that the "size" of the effect depends on how Lp(a) is measured, which is itself an unsettled area precisely because drugs like this one interfere with the test.

What to watch

KRAKEN moves muvalaplin forward as an oral option in a field dominated by injections, and it does so with a clean 12-week safety readout and a dramatic effect on the target [s1]. The meaningful threshold is still ahead: a trial large and long enough to show that driving Lp(a) down translates into fewer cardiovascular events, and that the pill stays safe over years rather than weeks. Until then, the right way to read an 85.8% reduction is as a promising laboratory result, not as proof of benefit.

This article is informational and does not constitute medical advice.

Sources

• [s1] Oral Muvalaplin for Lowering of Lipoprotein(a): A Randomized Clinical Trial. JAMA, 2025 (online 18 November 2024). https://doi.org/10.1001/jama.2024.24017
• [s2] A Study of LY3473329 in Adult Participants With Elevated Lipoprotein(a) at High Risk for Cardiovascular Events. ClinicalTrials.gov identifier NCT05563246, US National Library of Medicine. https://clinicaltrials.gov/study/NCT05563246